Parasaurolophus Walkeri

~r- ^ _92r"^" University Library

QE 862.D5P2517

Parasaurolophus Walkeri,

3 1924 004 594 580

CORNELL

UNIVERSITY

LIBRARY

UNIVERSITY OF TORONTO
STUDIES

GEOLOGICAL SERIES

No. 13: PARASAUROLOPHUS WALKERI, by W. A. Parks

THE UNIVERSITY LIBRARY: PUBLISHED BY
THE LIBRARIAN, 1922

lllnivcrsiti? of Toronto Studies

COMMITTEE OF MANAGEMENT

Chairman: Sir Robert Alexander Falconer, LL.D., K.C.M.G.,

President of the University

Professor W. J. Alexander, Ph.D.

Professor J. P. McMurrich, Ph.D.

Brig.-Gen. C. H. Mitchell, B.A.Sc., C.B., C.M.G , D.S.O.

Professor G. H. Needler, Ph.D.

Professor George M. Wrong, LL.D.

General Editor: W. S. Wallace, M.A.,

Associate Librarian of the University

PARASAUROLOPHUS WALKERI

A NEW GENUS AND SPECIES OF CRESTED TRACHODONT DINOSAUR

BY

William A. Parks, Ph.D., F.R.S.C.

Professor of Palaeontology
University of Toronto

c^s

1)6'

j if.^ f/

3/S

TABLE OF CONTENTS

PAGE

Discovery ... . . . . 5

General Description 5

Detailed Description . 7

Head 7

Vertebral Column 15

Pectoral Girdle .... 20

Fore Limb. . .... 22

Pelvic Girdle 27

Hind Limb 29

Musculature and Integument 31

Cornell University
Library

The original of this book is in
the Cornell University Library.

There are no known copyright restrictions in
the United States on the use of the text.

http://www.archive.org/details/cu31924004594580

PARASAUROLOPHUS WALKERI

a new genus and species of crested trachodont

dinosaur

Discovery

The specimen herein described was found in 1920 by an
expedition from the University of Toronto in the bad lands
of the Belly River formation on the Red Deer river, Alberta.
The actual discovery was made by Mr. L. W. Dippell on
the middle branch of the coulee below Sand creek at an
elevation of 125 feet (aneroid) above the river. The exca-
vation of the skeleton and its subsequent mounting in the
Royal Ontario Museum of Palaeontology was done under
the supervision of Mr. Levi Sternberg.

The skeleton was lying on its left side with some of the
bones of the right or upper side in an indifferent state of
preservation; in consequence, the mount has been prepared
to show the left or under side. The whole of the anterior
part including both fore limbs is well preserved, but the
tail and hind limbs with the exception of one femur are
wanting. The posterior ends of the ischia were also absent,
but have been restored in the mount. No other restoration
has been attempted.

General Description

The animal is evidently a representative of the Tracho-
dontidae and of the sub-family Saurolophinae or crested type
with footed ischia. It is characterized by a low-set and
heavy body and by a very remarkable type of crest which
seems to justify the creation of a new genus for its recep-
tion. Its affinities are undoubtedly with Saurolophus from
the Edmonton; hence, the proposed generic name, Para-
saurolophus. The specific name is given for Sir Edmund
Walker, Chairman of the Board of Trustees of the Royal
Ontario Museum.

The skeleton as mounted is shown by a carefully prepared
drawing (Plate I) and by a photograph (Plate II). The
anterior part is shown on a larger scale in Plate III. In all

6 General Description

these figures it is apparent that the left side of the head
has been sheared downwards. The drawing (Plate IV)
indicates the head as it would probably appear in true
lateral aspect. Some allowance must be made for the diffi-
culty involved in this reproduction; in fact, I am now of
the opinion that the supraorbital region has not been suf-
ficiently foreshortened vertically. The general size of the
animal is indicated by the following measurements:

GENERAL MEASUREMENTS

Tip of nose to anterior end of ilium 12 ft. 5 in.

Tip of nose to posterior end of ilium 15 ft. 9 J in.

Tip of nose to posterior rim of 41st centrum 16 ft. 9 in.

A comparison of numerous measurements indicates an
animal somewhat smaller than Saurolophus osborni, for
which Brown has estimated a total length of 32 feet.^

The resemblance to Saurolophus is very apparent in
the head in which the crest assumes an elongated spike-like
character rather than the plate-like appearance of such
forms as Corythosaurus and Stephanosaurus. The structure
of the premaxilla and of the external nares, however, suggests
affinities with Corythosaurus and the lachrymal bone seems
to be unique in its relationships. The size of the head,
also, disregarding the crest, is very much less than in Sauro-
lophus.

The body seems to be very like that of Saurolophus — •
massive and low-set. The pelvic girdle with the decurved
anterior process of the ilium and the greatly expanded pre-
pubis is strikingly similar to that of Saurolophus. The fore
limb shows more differences in proportion: the humerus
is almost exactly one-half the length of the femur, while in
Saurolophus the proportion is considerably greater. In
that genus, moreover, the radius and humerus are about
equal in length, while in Parasaurolophus the radius is con-
siderably shorter than the humerus.

•■Am. Mus. Nat. Hist., Vol. XXXI, Art. XIV, 1912; ibid., Vol. XXXII,
Art. XIX, 1913.

Parasaurolophus Walkeri 7

Detailed Description

HEAD

The left side of the head, which is exposed in the mount,
has been sheared downwards, thus increasing its apparent
depth: in the drawing (Plate IV) an attempt has been made
to place the parts in proper relationship. In general the
lower and lateral parts of the head conform to the usual
trachodont arrangement, but the superior margin is truly
extraordinary as it is continued backward in a gently convex
line to a greater distance than the total length of the head
proper.

This posterior prolongation or crest is a stout structure
of considerable thickness and depth the details of which
will be given later. It can scarcely be doubted that this crest
is analogous to that of Saurolophus, but it is much more
massive and directed less upwards than in that genus.

The general measurements of the head (I) are given
below with those of Saurolophus osborni (II) for comparison.

COMPARATIVE MEASUREMENTS OF HEAD

II

Tip of premaxillary to paraoccipital process

Tip of premaxillary to posterior of crest, measured along

superior surface

Ditto, measured in straight line

Postorbital notch to anterior point of premaxillary

Width of premaxillaries at widest part

Length in front of teeth

Length of quadrate

Length of mandible

Length of crest from notch behind the orbit to posterior

end

Width at postorbital rim, probably not exceeding

Height from inferior edge of articular in straight line at

right angles to inferior line of mandible to superior

margin

Height ditto to superior margin of squamosal

810 mm.

1200 mm

1733

675

170

380

270

500

280

350

640

1050

875

200

530

330

8 Head

It will be seen from these measurements that the head,
exclusive of the remarkable crest, is very much smaller
than in Saurolophus; in fact, the proportions are more like
those of Corythosaurus, in which the quadrate is of the same
length (280 mm.) and the mandible nearly the same
(669 mm.). In Stephanosaurus, likewise, the measurements
of the critical elements of the head are very similar — 27.5 mm.
for the quadrate, and 645 mm. for the mandible.

Premaxillary — The premaxillaries form the charac-
teristic duck-bill as in other genera of trachodonts: they are
excavated in wide narial depressions anterior to the nares
proper. This depression is floored throughout as in Corytho-
saurus, and there is a complete internarial septum. Whether
this septum is formed by ascending processes from the
premaxillaries or whether it is part of the nasal, it is im-
possible to say on account of the state of preservation of
the parts. The only suture visible is a median one in the
internarial bar. This can be traced forward a little in ad-
vance of the transverse line joining the anterior margins of
the two narial depressions. Here it forks, turns outwards,
and is lost. There is indistinct evidence that it turns back
again and cuts the anterior margin of the narial depression
about one-third of the distance across that opening. This
arrangement is distinctly different from that in Saurolophus,
where ascending processes from the premaxillaries separate
the nasals along the mid-line of the internarial bar.

Posteriorly also a distinct suture between premaxillary
and nasal cannot be deciphered with certainty; it is possible,
however, that a suture runs backward from the posterior
point of the narial opening, turns outwards and then for-
ward, and terminates at the deep suture above the maxillary.
If this line is really the premaxillary-nasal suture, it indi-
cates a condition somewhat like that of Prosaurolophus.
The most posterior point of this supposed suture is 460 mm.
from the middle point of the beak and the termination on
the premaxillary-maxillary suture 360 mm. from the same
point.

Parasaurolophus Walkeri 9

It must be understood that the above sutures are very
uncertain and that, as far as real evidence goes, the whole
of the superior surface of head and crest is composed of
premaxillaries. Maximum width of premaxillaries at duck-
bill, 170 mm.; width of contracted part behind duck-bill,
125 mm. ; width of narial opening, 80 mm. ; tip of premaxillary
to posterior suture with nasal (?), 460 mm.

Nasal — As already stated the sutures with the pre-
maxillary are problematical; it is reasonable, however, to
assume that the nasals occur, and if so, they must form
the major part of the superior surface of the head and of
the crest. A median sutural line can be observed superiorly,
and a deeply incised suture separates the nasal (and pre-
maxillary) from the maxillary and from the bone which
forms the lower half of the crest, and which is herein pro-
visionally regarded as the frontal.

Crest (Plate IV; Plate V, Fig. 1; text Figs. 1 and 2)—
If the interpretation above is correct the crest is formed of
nasals and f rentals; it fades imperceptibly into the superior
surface of the head proper; inferiorly it may be said to
terminate at a notch behind the orbit which it is proposed
to call the postorhital notch. The length from this notch to
the inferior-posterior end is 875 mm. and to the superior
margin 200 mm. The latter figure is doubtless too high as
it makes no allowance for the downward shearing which the
skull has suffered.

Posteriorly the crest is subquadrate in cross section, but
farther forward it is drawn out to a keel inferiorly; just
behind the postorbital notch it is sub-trapezoidal owing to
a widening of the inferior surface. The distal part is ex-
panded rather abruptly for a distance of about 20 mm. from
the end. The posterior surface of this expansion is strikingly
flat, is inclined downwards and backwards from the axis
of the crest, and measures 90 mm. transversely and 110 mm.
vertically. Laterally, this terminal expansion is sharply
defined against the nasal, but less distinctly against the
frontal. On the superior si^rface a median suture is plainly
discernible almost to the extreme end. The strongly marked

10

Head

lateral sutures (Plate V, Fig. 1) do not reach the extreme end,
but terminate in or near two deep pits. The inferior of
these pits is larger than the other and is situated a little more
anteriorly. This pit is so deep that it can not be separated
from its fellow of the opposite side by more than 25 mm.
The superior pit is smaller and seems to pass through the
terminal expansion as a sort of slit-like opening.

On the lateral aspect of the nasal, 170 mm. from the end,
is a deep pit which is continued backwards as a narrow
sulcus to the depression marking the anterior edge of the
swollen extremity. Above and posterior to this pit is a
well marked corrugated tuberosity 55 mm. in length. A
similar but smaller tuberosity lies below the sulcus at this
point.

A natural fracture crosses the crest 360 mm. from the
end: this break has enabled us to ascertain the character of
the cross section (Figs. 1 and 2). The crest is composed

FIG. 1 FIG. 2

Figure 1. Cross section of crest about 360 mm. from end, crushed as found.

One-third natural size.
Figure 2. Cross section of crest about 360 mm. from end, restored. One-third

natural size.

apparently of four tubes of very thin bone not more than
5 mm. thick except at the superior and inferior points of
junction. The lower line of junction, particularly, is con-
siderably thickened and drawn down to a hatchet-like edge.
The bones are coalesced as no sign of sutural union is visible.

Parasaurolophus Walkeri 11

The appearance of the cross section is indicated in
Figure 1, while Figure 2 is constructed to show the probable
cross section of the crest when free from distortion. If this
figure is correct the depth of the crest, 360 mm. from the
end, is 130 mm., the superior diameter, 90 mm., and the
inferior diameter, 85 mm.

In considering this remarkable structure, the question
naturally arises as to the character of these hollow bones
in life. Were they filled with organic tissues or were they
open air passages? The external nares seem to lead directly
into the upper pair of tubes, but nothing is known as to the
internal nares. This interesting question could probably
be answered by a longitudinal sawing through the skull,
but we have been unwilling to injure the specimen so seri-
ously.

Concerning the function of the crest itself, Brown is of
the opinion that in Saurolophus it bore a frill and that this
frill was probably continuous with a frill on the dorsal
vertebral spines. There can be little doubt that a similar
structure was present in Parasaurolophus, and we have
reason to believe that a muscular or ligamental union existed
between the crest and the dorsal vertebrae as will be more
fully considered later.

Maxillary — ^This bone shows the ordinary well-marked
sutures with the premaxillary and with the jugal: its relation
to the lachrymal, however, is open to question. The actual
suture with the premaxillary is 110 mm. long and with the
lachrymal (?) 40 mm. long. There are 31 strongly carinated
teeth visible in a length of 240 mm.

Lachrymal and prefrontal (Plate V, Fig. 2) — The identifi-
cation and relationship of these bones is extremely doubtful,
but it is certain that this genus is remarkably different from
any other trachodont in the arrangement of the bones of
this part of the head. A detailed drawing (Plate V, Fig. 2)
has been prepared of this part of the head to show the sutures
as far as they can be made out. Posterior to the maxillary,
and overlapping it, is a quadrangular to subtriangular bone
which it seems necessary to interpret as a lachrymal. Its

12 Head

suture with the jugal is distinct and it is separated from the
nasal (premaxillary?) by a deep pit which extends forward
above the maxillary and backward between the nasal and
the frontal almost to the end of the crest. Supero-posteriorly
there is a distinct suture with the bone behind, but infero-
posteriorly there is more doubt. While it can not be stated
with certainty there seems to be a small triangular bone
separating the infero-posterior margin of the lachrymal from
the jugal: this bone we are inclined to regard as the pre-
frontal.

If the interpretation above is correct the lachrymal in
Parasaurolophus is unique among the dinosaurs: in no other
case is it so far forward or is it excluded from the orbital rim.

The distinctness of the little bone which we have called
the prefrontal is very doubtful; it may be merely a process
of the jugal, as the occurrence of a suture along its inferior
border is open to question. As in the case of the lachrymal
the exclusion of the prefrontal from the orbital rim is re-
markable.

If this little bone is a mere process of the jugal, the
anterior part of the bone to be described later as the frontal
may be the prefrontal. No suture is visible above the orbit,
but the bone in this region is in a poor state of preservation,
and the absence of a suture cannot be stated with certainty.
This interpretation will doubtless appeal to many anato-
mists, as it leaves the prefrontal, at least, in the normal
position above the orbit and forming part of the orbital rim.

Frontal — This bone is unusually conspicuous as it forms
the whole of the lower half of the crest, and, apparently at
least, the whole of the supraorbital rim. As already stated,
however, the anterior part of this bone may possibly repre-
sent the prefrontal. The suture with the postfrontal is
distinct and as indicated in the drawing.

Postfrontal — This bone is normal in character and posi-
tion: its sutures with the frontal and with the squamosal are
distinct.

Jugal — This bone is of normal type: it is apparently not
in contact with the quadrate, but as the superior point of

Parasaurolophus Walkeri 13

the quadratojugal and contiguous parts of the jugal have
suffered some abrasion it is possible that the separation of
the jugal from the quadrate is less pronounced than shown in
the figure. The postorbital process is very slender, and there
is a very pronounced spur-like and somewhat elevated
point fitting into a recess at the suture with the frontal
(prefrontal ?). A deep pit occurs at the base of the upwardly
directed process or separate triangular bone which we have
considered, possibly at least, to be the prefrontal. Length
from supero-posterior to infero-anterior point, 237 mm.
Suture with quadratojugal, 95 mm.

Orbit — ^This aperture is oval with the greater diameter
strongly inclined backwards from the line of the teeth;
maximum length, 170 mm., maximum width, 105 mm.

Lateral temporal fossa — ^This vacuity is long and narrow :
195 mm. by 50 mm.

Quadrate — The quadrate is 280 mm. long and concave
posteriorly. The external width at the upper end is 45 mm. ;
just above the suture with the quadratojugal it is 65 mm.,
and at the narrow part above the lower end, 40 mm. It is
deeply excavated for the quadratojugal.

Quadratojugal — The visible part of this bone is very
narrow: its height is 95 mm. and its maximum external
width, 25 mm. As already stated it is somewhat doubtful
if it entirely separates the jugal from the quadrate.

Squamosal and paraoccipital process — The squamosal
forms a deep cotylus for the quadrate and sends down a
long process at the rear of the lateral temporal fossa: it
forms the external part of the paraoccipital process for
two-thirds of its length, the rest of this structure being
formed by the exoccipital as in other trachodonts.

Parietals — These bones meet in a narrow longitudinal
edge between the supratemporal fossae for a length of about
50 mm. Anteriorly the superior margin turns abruptly
upwards for 35 mm. to the inferior surface of the crest.
Here the bones flange outwards and backwards in over-
lapping sutures on the inferior surface of the frontals.
Laterally they sweep around and form the anterior margins of

14 Head

the supratemporal fossae. They seem to terminate laterally
in points between the frontals and postfrontals (i.e., within
and behind the postorbital notch, not shown in figure).

Supratemporal fossa — This opening is small and hidden
under the crest : owing to the sloping character of the margins,
exact measurements are difficult to make, but it may be said
to be about 95 mm. long and 50 mm. wide.

The mandible is of the usual trachodont type : the extreme
length is 640 mm., and the depth from the alveolar margin
to the inferior edge at the first tooth is 104 mm. From the
first tooth to the anterior point on midline is 200 mm.

Predentary — This bone is small and overlaps the dentary
with prominent suture. Width, 280 mm. ; length, 130 mm.
on midline from anterior point to line joining the posterior
prongs.

Dentary — The dentary is of the ordinary type, rather
strongly decurved anteriorly. The length on the inferior
margin from articulation with predentary to articulation
with surangular is 455 mm., and from the posterior prong
of the predentary to the same point, 390 mm. Thirty-one
teeth are visible in a distance of 240 mm. Still more may
occur, but if so they are hidden under the coronoid process.
There is evidence of at least three rows. Seven prominent
foramina occur on the external surface of this bone as shown
in the figure. The coronoid process is prominent and directed
slightly forward: it is 50 mm. wide at midlength. Its total
height cannot be ascertained as it is partially hidden under
the jugal, but it rises at least 190 mm. above the inferior
margin of the mandible and 90 mm. above the alveolar
margin.

Surangular — The upward extension of this bone lies
beneath the coronoid process with its margin almost parallel
with the posterior margin of the process: the bones are
separated by a deep sulcus. The suture with the dentary
runs irregularly down and back to the inferior edge of the
mandible.

Angular and articular — The sutures of these bones with
each other and with the surangular are not visible. From

Parasaurolophus Walker: 15

the inferior point of the suture between the surangular and
dentary to the posterior tip of the articular is 115 mm.

The strong flexure of the neck and the displacement of
the anterior vertebrae have revealed the foramen magnum,
which is small and apparently low in position. The other
features of the posterior aspect of the skull are not well
shown, but present nothing differing from the usual tracho-
dont arrangement.

VERTEBRAL COLUMN

The vertebral column consists of 13 cervicals, 17 dorsals,
7 sacrals, and 6 caudals preserved. The vertebrae of the
sacral region are not sufficiently exposed in the mount to
state definitely how many are coalesced. The actual con-
ditions are shown in the photograph (Plate VI). It is ap-
parent that the vertebral centra shown in Plate I are very
largely restored.

Cervicals (Plate VII, Figs. 1, 2, 3)— The total length of
the thirtefen cervicals measured along the line of the dia-
pophyses is about 1190 mm. A more accurate measure-
ment is 670 mm. along the ventral side from the posterior
rim of the 7th to the posterior rim of the 13th vertebra.

The cervical centra are strongly opisthocoelous. Cervical
ribs are present, but owing to the neck having been forced
between the maridibles the anterior cervical ribs do not
appear. The most striking feature of the neck is the re-
markably strong and prominent postzygopophyses.

The atlas is not well exposed: the length of the neural
arch is 70 mm. and the maximum diameter across the two
arches is 110 mm.

The axis carries the usual prominent plate-like neural
spine, which is 85 mm. long on the midline. The total
length to the posterior point of the postzygopophysis is
145 mm., and the width between the external surfaces of
the two zygopophyses is 80 mm. The postzygopophysial
facette is 38 mm. long.

The first apparent neural spine after that of the axis is
the ninth, which is only about 10 mm. high. Posteriorly the

16 Vertebral Column

spines increase in size as follows: 10th, 35 mm.; 11th, 40 mm.;
12th, 50 mm.; 13th, not measurable.

The first measurable cervical rib is the fifth which is
long and slender, 105 mm. from diapophysis to tip. The
sixth is shorter, and less pointed: its length is not measur-
able. The seventh to tenth inclusive are flatter, broader,
square at the end, and of about equal length — 90 mm., 100
mm., 90 mm., 85 mm. The eleventh to thirteenth are more
pointed. The eleventh is 90 mm. long and the thirteenth
120 mm. The twelfth is not well preserved.

The diapophyses of the cervical vertebrae are well
developed: that of the axis is small and uncertain, that of
the 7th vertebra is 60 mm. long.

The postzygopophyses are very strongly developed.
From the eighth to the thirteenth they are curved down-
ward and carry a well-marked facette. The curvature of
the postzygopophyses ceases abruptly with the first dorsal.
The lengths measured along the curve are approximately as
follows : 3rd, 50 mm. ; 5th, 70 mm. ; 6th, 80 mm. ; 8th, 85 mm. ;
9th, 85 mm.; 10th, 90 mm.; 11th, 90 mm.; 12th, 100 mm.;
13th, 110 mm.

Plate VII shows the eighth, ninth, and tenth cervical
vertebrae from the left lateral, dorsal, and ventral points of
view.

Dorsal vertebrae — There are apparently 17 dorsal vertebrae
carrying ribs. The centra are decreasingly opisthocoelous
posteriorly: the size of the centra is not measurable, but the
posterior ones are approximately 80 mm. long, 150 mm.
high, and 72 mm. wide.

The first dorsal spine is distinctly more pointed than the
last cervical, and the spines gradually increase in length
posteriorly: they are rather variable in shape, and the
posterior ones are closely set and both broad and long. The
fifth, sixth, and seventh show a remarkable modification
which will be referred to later.

The postzygopophysis of the first dorsal is distinctly
shorter than that of the last cervical, being only about
80 mm. long as compared with 110 mm. The second dorsal

Parasaurolophus Walkeri

17

postzygopophysis is only 50 mm. and the 16th is quite small.

The diapophyses are prominent, long, and somewhat
variable in shape: measurements are given in the table
below.

The ribs are very heavy when compared with those of
such genera as Corythosaurus or Kritosaurus. The fourth,
fifth, and sixth ribs show distinct evidence of injury during
the life of the animal. There is a distinct fracture observ-
able in the fourth, and the fifth and sixth show swollen
coalescences of broken parts.

MEASUREMENTS OF CERTAIN ELEMENTS OF THE DORSAL

REGION

Dorsal

Neural

spine

Diapophysis

Rib

vertebra,

mm

.

mm.

length, mm.

No.

Leng;th

Width at
top

Length

Width

1

160

130(?)"

360

2

165

120

from

590

3

170

120

centrum

840

4

135

970

5

180

135 ■

1050

6

140

120

1130

7

140

110

1140

8

295

55

95

125

1115

9

290

85

100

100

918

10

305

85

95

from

80

685

11

305

85

100

.base of

400+

12

300

100

110

neural

280+

13

285

95

120

spine

275

14

270

80

150

245

15

275

75

120

198

16

280

80

120

105

17

180(?).

present but

indetermin-

able

It will be observed from the above table and from a
glance at the figure that the sixth neural spine is unusually

18 Vertebral Column

wide and straight; that the fifth is pointed and inclined
backwards so as almost to touch the sixth; and that the
seventh, on the other hand, is inclined forward and is ap-
parently in actual contact with the tip of the sixth. The
sixth spine, in particular, and to some extent the seventh,
bears at its tip a discoidal expansion or co-ossified separate
bone about 100 mm. in diameter and 20 mm. thick. This
pad of bone is rough and without distinct surface. At first
we were inclined to regard it as the result of injury, more
particularly as there is distinct evidence in the broken ribs
already referred to of the animal having suffered an accident.
There is another explanation, however, which appeals to
us strongly, although at first one is tempted to regard it as
fanciful. May it not be that the crest of the head was
attached to this pad of bone by muscles or ligaments? It
has already been suggested that the crested trachodonts
bore a frill which extended down the neck and into the
dorsal region. It is not hard to imagine that we have here
an extreme case of the same sort of thing. If the abnormal
flexure of the neck were removed and the head and crest
placed in natural position, the end of the crest would be
considerably above and in advance of this pad of bone on
the sixth dorsal vertebra. That such connection between
the crest and the vertebrae existed in life is further suggested
by the very considerable difference between the seventh
and eighth neural spines, the latter of which is more than
twice as long as the former. Again, the dislocation of the
column between the seventh and eighth vertebrae suggests
an association of the anterior part with the head and im-
plies a participation in the series of events in connection
with the abnormal flexure of the neck whether before or
after death.

Sacrum — It is quite impossible in view of the condition
of the mount to state accurately the number of sacral verte-
brae. There are at least six thoroughly fused centra and
a seventh, partially united, which may be regarded as a
sacro-caudal. The six anterior neural spines are high and
wide like those of the posterior dorsals, but the seventh is

Parasaurolophus Walkeri

19

narrower and resembles those of the undoubted caudals.
The chief reason why this vertebra is regarded as a sacro-
caudal is the length of the centrum, which is 120 mm., as
compared with 90 mm. for the first true caudal.

MEASUREMENTS OF SACRAL VERTEBRAE

Sacral vertebra

Neural spines

Diapophysis

No.

Length, mm.

Width, mm.

Height from base of

neural spine, mm.

1

310

90

140

2

310

90

160

3

305

80

150

4

310

85

150

5

330

80

145

6

370

200

7

405

65

170

Caudal vertebrae — Only four caudal vertebrae are perfectly
preserved, but parts of the fifth and sixth are present; these
have been restored in the mount, but are not referred to in
the measurements given below.

MEASUREMENTS OF CAUDAL VERTEBRAE

Caudal vertebra
No

Length of
centrum, mm.

Neural spine

Diapophysis
Length, mm.

Length
mm.

Width
mm.

From
base of sp.

From
centrum

1
2
3
4

90
80
80
80

•415
410
400
390

60
60
60
50

150
160
140
125

105
80

72

The 41st centrum, the only one measurable, is 80 mm.
long, 150 mm. high, and 72 mm. wide.

20

Pectoral Girdle

PECTORAL GIRDLE

The girdle consists of the scapulae and coracoids as usual
in the dinosaurs. Sternal bones were not found.

Scapula — This bone is unusually heavy and wide in the
blade and it is firmly united with the coracoid. The photo-
graph (Plate III) and the comparative measurements will
serve to define it.'

COMPARATIVE MEASUREMENTS OF SCAPULA

Parasaurolophus

Corythosaurus

Kritosaurus

Saurolophus

walkeri

casuanus

mcurvimanus

osborni

Total length

940

890

776

900

Maximum width

of blade

248

200

189

220

Width at articu-

lar end

235

214

Length of glenoid

cavity

125

110

Width of glenoid

cavity

75

64

It will be observed that the scapula is of greater length
than that of Saurolophus: it is surpassed only by that of
Claosaurus annectens (970 mm.). It shows a greater width
of blade than any other form of which I can find a description.

Coracoid (Fig. 3) — Neither of the coracoids is well pre-
served: the left is in better condition and is shown, with
considerable adjustment, in Figure 3. The bone is firmly
united with the scapula and the foraminal notch is closed
as in all trachodonts with the exception of Kritosaurus. The
following rather uncertain measurements indicate the size
of the bone. The glenoid cavity has a transverse width of

^Owing to disregard of an erratum slip accompanying Barnum Brown's
paper on Saurolophus osborni, some of the measurements quoted in University
of Toronto Studies, Geological Series, No. 11, are transposed. In the table on
page 41 of that study, 11 should read Saurolophus osborni, III Trachodon mirabilis,
and IV Claosaurus annectens.

Parasaurolophus Walkeri

21

80 mm. and a vertical width of 75 mm. From the infero-
anterior point to the posterior of the articulation with the
scapula is 185 mm. and from the margin of the glenoid

Figure 3. Left coracoid, external view. One-third natural size.

cavity to anterior edge of the bone, 105 to 110 mm. The
upper portion of the twisted anterior margin is much more
lobe-like than in the case of Kritosa-iirus.

FORE LIMB

Humerus (Figs. 4 and 5) — -This bone is of the ordinary
type, but it is remarkable for its short length and heavy
character. This is the more noticeable in view of the great
length of the scapula. The figures and the comparative
measurements given herewith require little comment. The
great radial crest with its upward aspect is rather striking.

COMPARATIVE

MEASUREMENTS

OF HUMERUS

Parasaurolophus
walkeri

Kritosaurus
incurvimanus

Saurolophus
osborni

Length

Width across radial crest

520
160
310

255

630
170

214

610
310

Girth of shaft

22 Fore Limb

COMPARATIVE MEASUREMENTS OF HVMERUS—Conlinued

Parasaurolophus
walkeri

200
150

70(?)

88
84
40
40

Kritosaurus
incLirvimanus

100

80

82
78
41
31

Saurolophus
osborni

Width across inner and outer
tuberosities

Width across condyles

Thickness of outer condyle (ant.-
post)

Thickness of inner condyle (ant.-
post)

Thickness of head

Thickness ol inner tuljerosity . . . .

Thickness of outer tulierosity . . . .

s 1

\ ]

f f

sJ, '

'4, ; ,A*

FIG. 4 FIG. .')

Figure 4. Left humerus, external \-iew. One-sixth natural size.
P'igure 5. Left humerus, internal view. One-sixth natural size.

Ulna (Figs. and 7; Plate VI 11, Figs. 1 and 2)— This
bone requires little comment: it is quite the same as in

Parasaurolophus Walkeri

23

other trachodonts. The proximal parts of both bones are
rather badly crushed, rendering exact measurements difficult.
The only significant features are indicated in the comparative
measurements given below.

COMPARATIVE MEASUREMENTS OF ULNA

Parasaurolophus
walkeri

Kritosaurus
incurvimanus

Sauroiophus
osborni

.560
180

610
1.58

680

Minimum girth of shaft

Radius (Plate VI 11, Figs. 1 and 2) — This bone has the
usual form of a long thin shaft expanded at both ends.

m

il M

Wl

FIG. 6 FIG

Figure 6. Left ulna, intero-anterior view. One sixth natural size.
Figure 7. Left ulna, externo-posterior view. One-sixth natural size.

24

Fore Limb

Neither bone was in the best of condition and, in conse-
quence, a small allowance may be necessary in the figures
given below: those for the left bone are the more reliable.

COMPARATIVE MEASUREMENTS OF RADIUS

Parasaurolophus
walkeri

Kritosaurus
incurvimanus

Saurolophus
osborni

Length . .

right left
485 496

152 152
160 148
140 146

555
136

620

Girth of shaft, one-third from dis-
tal end . ....

Girth, one-third from prox. end.. .

Carpals — In both limbs a small carpal was found lying
on the distal posterior edge of the ulna: a second carpal was
not observed.

Manus (Plate VIII, Fig. 3; Plate IX, Figs. 1, 2, 3, and 4)—
As in the case of the other elements of the fore limb the
manus is short. Metacarpals III and IV are closely associ-
ated and longest; metacarpal II is somewhat shorter, but
nevertheless closely associated with III and IV. Metacarpal
V is much smaller and divergent. The phalangeal formula
of trachodonts was discussed in a former paper.^ Although
the terminal phalanx of Digit V is lacking, there seems no
reason to doubt that the formula for Parasaurolophus is the
same as that for Kritosaurus as follows :

Digit II with three phalanges, the third a pointed hoof.

Digit III with three phalanges, the third a broader hoof.

Digit IV with three phalanges, no hoof.

Digit V with four phalanges, the terminal a small ovoid

bone (this last phalanx is not preserved).

As in the case of Kritosaurus, phalanges IP and IIP are

triangular with the thin edge directed inwards: in the case

of the latter of these bones, the triangular character is not

so marked as in Kritosaurus.

' Parks, University of Toronto Studies, Geol. Ser., No. 11, 1920.

Parasaurolophus Walkeri

25

Both feet are preserved intact with the exception of
phalanx V"* in both cases. While the bones are to some
extent imperfect and deformed, there is, nevertheless, a
certain amount of difference between the two feet which
is not due to accidents after entombment. The most striking
of these differences is the abnormally narrow distal end of
metacarpal III in the left foot.

It has been thought better to illustrate the manus fully
rather than to give detailed descriptions of the individual
bones. The plates and the comparative measurements given
below should suffice.

COMPARATIVE MEASUREMENTS OF METACARPALS

•il

en

a,

to

:3
u

D
oi

to

3
C

'>
11
3
o

a

c
o .t;

oj w

>- o

O nj

CD J3

3 ^
cn o

u

i4

u '3

H S

-.3 3 5

U

O 3

b.s

5 a

Q. CD

Metacarpal II, length

Metacarpal II, max. diameter
proximal end

Metacarpal II, max. diameter
distal end

Metacarpal III, length

Metacarpal III, max. dia-
meter proximal end ....

Metacarpal III, max. dia-
meter distal end

Metacarpal IV, length

Metacarpal IV, max. dia-
meter proximal end . . . .

Metacarpal IV, max. dia-
meter distal end

Metacarpal V, length

Metacarpal V, max. diameter
proximal end

Metacarpal V, max. diameter
distal end

mm.

mm.

mm.

mm.

173

250

185

200

44

39

45

37

208

330

226

220

54

49

60

50

195

330

225

215

61

68

57

51

90

130

94

75

48

52

32

35

mm.
245

mm.
220

265

280

26 Fore Limb

comparative measure.mexts of phalanges

Parasaurolophus
walkeri

Kritosaurus
incurvimanus

Phalanx II

IP,

IP,

Phalanx
Phalanx

Phalanx

Phalanx
Phalanx

Phalanx IV

Phalanx
Phalanx
Phalanx

Phala

Phalanx
Phalanx

length

proximal width .

distal width. . . .

length

width

length

proximal width. .

width of hoof . . .
IIP. length

proximal width. .

distal width

IIP, length

width

IIP, length

proximal width. .

width of hoof. . .

length

proximal width. .

distal width

length

width

length

width

length

proximal width. .

distal width

length

proximal width. .

distal width

V, length

width

\'\ length

1V\

Iv^

: V=,

mm.

65

41

45

17

31

58

34

37

38

58

55

15

40

51

46

52

54

48

42

14

33

11

17

48

39

34

30

28

27

25

mm.

73

45

42

18

25

64

37

53

57

51

47

21

34

52

45

58

66

52

53

18

34

15

21

60

60

32

18
10

The relatively large size of the scapula has already been
referred to: equally noteworthy is the relatively short fore
limb in all its elements. The humerus is considerably
shorter than in Saurolophus osborui, but somewhat longer

Parasaurolophus Walkeri 27

than in Trachodon and Claosaurus. The ulna is very much
shorter than in Saurolophus, and only in Trachodon mirabilis
is a shorter one recorded. The same relationship holds for
the radius. The radius is shorter than the humerus, while
in Saurolophus it is slightly longer {vide preliminary descrip-
tion, p. 6).

If Saurolophus osborni be regarded as the nearest relative
of the present species, a most striking difference is shown by
the manus. Saurolophus appears from Brown's figures to
have the longest manus of any trachodont; on the other
hand, Parasaurolophus has the shortest.

Adding together the lengths of humerus, radius, and
metacarpal III, we obtain the following figures for the lengths
of the fore limbs exclusive of phalanges:

mm.

Parasaurolophus walkeri 1224

Kritosaurus incurvimanus 1411

Saurolophus osborni 1560

Trachodon mirabilis 1161

Claosaurus annectens 1345 (metacarpal II instead of III)

Hypacrosaurus altispinus 1545

It will be observed that Saurolophus has the longest fore
limb and that Parasaurolophus has the shortest with the
exception of Trachodon mirabilis. These relationships would
not be greatly changed even if allowance were made for
difference in size of the animals. While the fore limb of
Parasaurolophus is short, it is relatively stout and is sup-
ported by a scapula of extraordinary size. May it not be
that this arrangement is a direct consequence of the supposed
attachment of the crest to the vertebral column? Short
fore limbs would seem to be a necessity in order to permit
the animal to lower its head while eating.

PELVIC GIRDLE

The pelvic girdle is distinctly of the type described by
Brown for Saurolophus} It is deep and massive with
decurved anterior process on the ilium and with expanded

Mm. Mus. Nal. Hist., Vol. XXXIII, Art. XIX, 1913.

28

Pelvic Girdle

prepubis. A peculiar feature is that the ilium seems to
form an overlapping union with the pubis instead of abutting
against it as in other forms. After a careful examination,
I am of the opinion that this arrangement is not due to
deformation.

Ilium (Fig. S, Plate \'I) — This bone is ^'ery massive with
stout ischiac peduncle above which the overhanging shelf
is sharply defined and comes low down. The pubic peduncle

Figure S. Left pelvic girdle, external view. One-twelfth natural size.

is unusually extended anteriorly, tlecurved at its anterior
end and fastened to the pubis Ijy an externally o\-erlapping
joint. Even if this manner of union is due to distortion,
there can be no doubt as to the quite unique character of
the anteriorly extended pubic peduncle.

Ischium (Fig. 8, Plate VI) — This bone seems to be less
massive than the ilium and to present nothing of particular
interest. Only the proximal portion is preserved, but the
distal end has ficen restored with a "foot" in the mount.

Pubis (Fig. 8, Plate YI) — The prepubis is generally like
that of Saurolophus, but has an outline sufficiently different
to distinguish it. The anterior end is much less decurved
and the narrow portion is relatively wider. It differs greatly

Parasaurolophus Walkeri

29

from the prepubis of Kritosaurus, in which the upper and
lower margins of the blade-like part are approximately
parallel. The peculiar union with the ilium has been referred
to above.

The postpubis is rather short and flattened towards the
extremity. It seems to lie farther from the ischium than
usual. The pubic notch is open. The acetabulum is elon-
gated longitudinally.

COMPARATIVE MEASUREMENTS OF PELVIC GIRDLE

il

3
j3

c

-§.2

n

a

1-

O i_,
1- o

2 S

a, a

J5 s

H 6

U 3

Ilium, length in straight line
length along superior

margin

length of anterior pro

cess

Ischium, length

" width 300 mm. from

proximal end

girth at this point . .

Pubis, total length

width of blade of pre

pubis

preacetabular border

to anterior end.

" iliac union to anterior

end of prepubis.

1015

1160

1013

1030+

1120

445

480

1200

1026

1090

110

280

894 (?)

1150

1038

630

260

310

163

200

516

590

519

360

636

1150

220
450

HIND LIMB

The only part of either hind limb preserved is the left
femur.

Femur (Fig. 9, Plate VI) — This bone is of the ordinary
type, and is sufficiently defined by the figures and the com-

30

Hind Limb

parative measurements given below. The less prominent
head in Figure 9, compared with that shown in Plate VI,
Fig. 1, is due to the fact that the bone is thrown inwards
distally, thus bringing the head into view. The figure is
drawn in true projection.

COMPARATIVE MEASUREMENTS OF FEMUR

° 'S

3

c
o .—

o ^

x;

Z ^

3 ra

o •-•

^ .5

o E

trt L.

>- o

Oh a

2 -^

!/3 O

W .5

e^l

U 3

Length, outer condyle to proxi
mal end

Width across trochanters

Width of inner condyle

Width of outer condyle

Thickness at condyles

Width of shaft above condyles .

Thickness of shaft above con
dyles

Girth of shaft above condyles.

Width of head, antero-postero .

Length of 4th trochanter

Height of 4th trochanter

Thickness across head to mid-
line between trochanters

1032

1150

1045

1040

230

209

275

292

285

262

180

210

130

110

410

185

290

100

265

1170

It will be observed that the lengths of the different
femora do not greatly differ. The other measurements are
not very reliable, as crushing in many cases seriously affects
the proportions of different parts. For instance, it is un-
likely that the figures given above for the widths of the inner
and outer condyles are correct. Why should the inner
condyle exceed the outer in width in Kritosaiirus and the
opposite condition maintain in Parasanrolophus?

While the pelvic girdles of trachodonts are distinctive for
the different genera, it has not yet been established that the
same holds true for the bones of the hind limb, which seem

Parasaurolophus Walkeri

31

to be essentially so similar that only the greatest of detail
will serve to differentiate the genera a!nd species. The fore
limb, on the other hand, indicates faithfully the specializa-
tion of the animal and is, in consequence, of greater value in
the work of identification.

EPIDERMIS AND MUSCULATURE

Many ossified tendons were found in the course of pre-
paration, but they were so broken and scattered that it was

Figure 9. Left femur, external view. One-twelfth natural size.

impossible to save them. It was noticed that these tendons
were not in close contact with the neural spines, as is
generally the case.

32 Epidermis and Musculature

Impressions of the skin also were found in several parts
of the body: they all exhibited the usual tuberculated
structure, but in no case revealed any of the acorn-like
elevations seen in other trachodonts. Apparently the skin
was uniformly tuberculated, but it is impossible to say
that this condition maintained over the whole animal.

u 11
3

O

-a
c .

6 6

w

H

<!
J

3-13

rt^

c

>

OJ

-C

tu

V

rt

\~

d .

T3.-H

y>< /

O
<

a
o

a.

\ "*

rurdsauinUiphiis walker i.
Figure 1. Detail of end (if crest. Alimit 0/20 natural size.
Fiqaire 2. Detail nf su|>ra-].ireiirliital region. Aliout 2/i-! natural size.
F frontal; |, JLigal; L, laehr\-nial (,?); M, maxillary; X, nasal; P, prefrontal {?).

o
<

Plate VII

Figure 1. Left lateral view.

Figure 2. Dorsal view.

Figure 3. Ventral view.
PaniSiiiirflhiphiis i^HiUicri. Eighth, ninth, and tenth cervical vertebrae. About
l/'3 natural size.

^l.ATE VIII

^"JJ

P:ii{isiiiiiiili)f}lnis 7^'iilkcri.
Figure 1. I I li lailius iml ulna, cxtLTno-po>tcrior \-ie\v.

Miout 1 7 natural size.
Fit;ure '2. I i U i ulnis ind ulna, intero-anterior \-iew.

Aliiiut 1 7 natural size.
Fiyure 3. l.ill manus, ilissociated, anterior \-ie«-,

.\bout 1 y natural size.

Plate IX

;««»

Figure
Figure
Figure

1<

Parasaiirohiphus 'a-dlkcri.
lit man us, anteridr view. Aliout 1/7 natural size.
I.tft uiaiius, anterior view. AVjout 1/7 natural size.
LVft manus, posterior view. About 1/7 natural size.
Ki.'ht manus, posterior view. About 1/7 natural size.

UNIVERSITY OF TORONTO STUDIES
Geological Series

No. i: The Huronian of the Moose River basin, by

W. A. Parks ^0.50

No. 2: The Michipicoten iron ranges, by Professor A. P.

Coleman and W. A. Willmott i .00

No. 3: The geology of Michipicoten Island, by E. M.

BURWASH i.oo

No. 4: The stromatoporoids of the Guelph formation in

Ontario, by Professor W. A. Parks i.oo

No. 5: Niagara stromatoporoids, by Professor W. A.

Parks i .00

No. 6 : Silurian stromatoporoids of America, by Professor

W. A. Parks i .00

No. 7: Ordovician stromatoporoids, by Professor W. A.

Parks i.oo

No. 8: A Cervalces antler from the Toronto Interglacial,

by Professor B. A. Bensley 0.25

No. 9: Palaeozoic fossils from a region southwest of

Hudson Bay, by Professor W. A. Parks i.oo

No. 10: Mineralogy of the H. B. Mine, Salmo, B.C., by

Professor T. L. Walker 0.50

No. 1 1 : The osteology of the trachodont dinosaur Krito-

saurus incurvimanus, by Professor W. A. Parks i.oo

No. 12: Contributions to Canadian Mineralogy, 1921 i.oo

No. 13: Parasaurolophus Walkeri, a new genus and species

of crested trachodont dinosaur, by Professor

W. A. Parks i .00

w

PAM PHLET BINDER

^^3 Syracuse. N. Y
;^^^ Stockton, Calif